Regression-Based Approaches for Remaining Useful Life Prediction of Proton Exchange Membrane Fuel Cells

Résumé fourni par la source

This study focuses on regression-based approaches for the prediction of the Remaining Useful Life (RUL) of Proton Exchange Membrane Fuel Cells (PEMFCs). By employing both statistical regression models and advanced machine learning regression techniques, the research aims to evaluate their effectiveness and comparative performance in capturing degradation patterns. The dataset, consisting of the IEEE PHM 2014 Fuel Cell Dataset with sensor measurements from PEMFCs under static and dynamic load conditions, was used for training and testing the models. Experimental results demonstrate that tree-based ensemble methods, particularly Random Forest and Gradient Boosting, outperformed traditional regression techniques by achieving significantly lower RMSE values. These findings highlight the potential of regression-based modeling as a robust tool for PEMFC health management in automotive and energy applications.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.